This project will design and field the Third Malaysian Family Life Survey (MFL-3) in 1999. It will be based on the First and Second Malaysian Family Life Surveys (MFLS-1 and -2), fielded in 1976 and 1988, respectively. All MFLS-2 respondents will be selected to be interviewed in MFLS-3, as will many of their adult children; in addition, the survey will include a new sample that will be representative of Peninsular Malaysia. MFLS-3 will collect information from up to three generations of adults from the same family and will provide the third consecutive cross-sectional data set for Malaysia. The three MFLS surveys together enable researchers to examine the effects of family background on socioeconomic behavior and demographic outcomes, and to study demographic and socioeconomic change over time, over the life-course, and across generations. Information on the same individuals will be available at up to three points in time: 1976, 1988, and 1998. The MFLS-3 will include strong longitudinal, intergenerational, and intra- family components. The longitudinal component includes repeated nationally-representative cross-sections and repeated interviews with the same respondents, and will allow researchers to analyze the patterns of demographic change over time and over the life course. The intergenerational component permits the study of change across generations in demographic behavior and outcomes, and allows researchers to control for the effects of family background. The intra-family component is a consequence of including multiple adult children from the same family as respondents in the survey, permitting analyses that control for unobserved family characteristics using family fixed-effects or random-effects models. RAND will collaborate with the National Population and Family Development Board of Malaysia (LPPKN), which also collaborated on the MFLS-2.